Research

Causal decomposition of Black–white disparities in NIH's selection of proposals for panel discussion

• In collaboration with Elena Erosheva, Carole Lee
• We extend causal decomposition methodology to identify drivers of Black–white disparities in NIH R01 panel discussion selection and evaluate policy-relevant interventions to reduce those disparities.
• Student Paper Competition Winner (ASA SSS/GSS/SRMS 2026)
• Paper in review

Hypothesis testing for functional causal discovery

• In collaboration with Elena Erosheva and Fan Xia
• We identify a gap in statistical guarantees for functional causal discovery and introduce a unified, test-based inferential framework for bivariate causal direction.
• Presented at American Causal Inference Conference (ACIC)
• Talk at The Western North American Region of The International Biometric Society (WNAR)
• Paper in review

Statistical Causal Discovery via Subsampling

• In collaboration with Elena Erosheva and Fan Xia
• We introduce Statistical Causal Discovery via Subsampling (SCDS), a power-based framework that connects statistical inference with causal direction detection and remains informative even under model misspecification.
• Paper in preparation